Someone who believes a Minnesota officer used excessive force usually assumes the case turns on whether the force was reasonable. It often does not. It turns on a prior question that has nothing to do with what happened to them: whether, before the incident, some earlier court had already held that force like this, in circumstances like these, was unconstitutional.
That is qualified immunity. It is not a defense on the merits. It is a rule about the state of the case law on the day the officer acted, and in the Eighth Circuit — which governs every federal civil rights case filed in Minnesota — it ends claims that never reach a jury, including claims a district judge had already concluded a jury should hear. The case discussed below is one of them. Any honest description of this doctrine has to say that out loud.
Where the doctrine comes from
The modern rule dates from Harlow v. Fitzgerald, 457 U.S. 800 (1982), which stripped the old “good faith” inquiry of its subjective element. Government officials performing discretionary functions, the Court held, “generally are shielded from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.” Id. at 818. The stated purpose was to “permit the resolution of many insubstantial claims on summary judgment.” Id.
Two features follow from that design and matter enormously in practice.
First, the officer’s actual state of mind is irrelevant. A defendant who acted in bad faith still gets immunity if the law was not clearly established; a defendant with pure motives still loses immunity if it was.
Second, immunity is from suit, not merely from liability. Mitchell v. Forsyth, 472 U.S. 511 (1985), held that “[t]he entitlement is an immunity from suit rather than a mere defense to liability; and like an absolute immunity, it is effectively lost if a case is erroneously permitted to go to trial.” Id. at 526. From that the Court drew a procedural consequence: “a district court’s denial of a claim of qualified immunity, to the extent that it turns on an issue of law, is an appealable ‘final decision’ within the meaning of 28 U.S.C. § 1291 notwithstanding the absence of a final judgment.” Id. at 530. A defendant who loses on immunity can therefore take the question to the court of appeals before trial. A plaintiff who loses on immunity has no equivalent mid-case appeal.
“Clearly established” is a demand for specificity
The heart of the doctrine is not the phrase “clearly established.” It is how narrowly courts are required to define the right.
Anderson v. Creighton, 483 U.S. 635 (1987), set the frame: “The contours of the right must be sufficiently clear that a reasonable official would understand that what he is doing violates that right.” Id. at 640. The Court explained why it would not accept a general statement: otherwise plaintiffs “would be able to convert the rule of qualified immunity that our cases plainly establish into a rule of virtually unqualified liability simply by alleging violation of extremely abstract rights.” Id. at 639.
The Court has repeated that instruction with growing impatience. In Ashcroft v. al-Kidd, 563 U.S. 731 (2011), it said “[w]e have repeatedly told courts . . . not to define clearly established law at a high level of generality,” id. at 742, and that while “[w]e do not require a case directly on point, . . . existing precedent must have placed the statutory or constitutional question beyond debate,” id. at 741. Absent controlling authority, what is needed is “a robust ‘consensus of cases of persuasive authority.’” Id. at 742 (quoting Wilson v. Layne, 526 U.S. 603, 617 (1999)).
Mullenix v. Luna, 136 S. Ct. 305 (2015) (per curiam), put the point in one sentence: “The dispositive question is ‘whether the violative nature of particular conduct is clearly established.’” Id. at 308 (quoting al-Kidd). And District of Columbia v. Wesby, 138 S. Ct. 577 (2018), tightened it further: “It is not enough that the rule is suggested by then-existing precedent. The precedent must be clear enough that every reasonable official would interpret it to establish the particular rule the plaintiff seeks to apply.” Id. at 590.
The practical translation is blunt. In White v. Pauly, 137 S. Ct. 548 (2017) (per curiam), the Court faulted a court of appeals because “[i]t failed to identify a case where an officer acting under similar circumstances as Officer White was held to have violated the Fourth Amendment.” Id. at 552. That is what a plaintiff is expected to produce: a prior decision with facts close to their own.
There is an escape valve, and it is real but narrow. Hope v. Pelzer, 536 U.S. 730 (2002), holds that “officials can still be on notice that their conduct violates established law even in novel factual circumstances,” id. at 741, and Wesby acknowledges “the rare ‘obvious case’” where no closely analogous decision is needed. But a plaintiff who is relying on obviousness rather than precedent is relying on the exception.
What this looks like in the Eighth Circuit
Kelsay v. Ernst, 933 F.3d 975 (8th Cir. 2019) (en banc), is the clearest illustration available. It arose in Nebraska rather than Minnesota, but it is an en banc decision of the circuit whose law binds Minnesota’s federal courts. It is worth stating the facts as the court itself stated them, viewed in the light most favorable to the plaintiff.
Melanie Kelsay, five feet tall and about 130 pounds, was at a public pool. Her friend was arrested. She was upset, walked toward her daughter, and ignored a deputy’s instruction to “get back here.” The deputy “placed Kelsay in a bear hug, threw her to the ground, and placed her in handcuffs.” Id. at 978. She lost consciousness momentarily and was later “diagnosed with a fractured collarbone.” Id. at 978-79. The district court denied qualified immunity, reasoning that she “was not using force or actively resisting arrest, and posed no danger to anyone.” Id. at 980.
The en banc Eighth Circuit reversed. Not because the force was reasonable — the court did not decide that. Because “[i]t was not clearly established in May 2014 that a deputy was forbidden to use a takedown maneuver to arrest a suspect who ignored the deputy’s instruction to ‘get back here’ and continued to walk away from the officer.” Id. The plaintiff’s cases involved suspects who were compliant, or passively resistant, or standing still — none involved a suspect who ignored a command and walked away. “None of Kelsay’s authorities ‘squarely governs the specific facts at issue.’” Id. (quoting Kisela v. Hughes, 138 S. Ct. 1148, 1153 (2018) (per curiam)).
Read that again in terms of what it means for the next plaintiff. A broken collarbone, a district judge who concluded on this record that the plaintiff posed no danger and that a jury should hear the case — and the claim ended because the difference between standing still and walking away had not been litigated to judgment before. This is the doctrine functioning exactly as written.
Immunity is not automatic. Ludwig v. Anderson, 54 F.3d 465 (8th Cir. 1995), reversed a grant of qualified immunity to “two St. Paul, Minnesota, police officers,” id. at 467, holding that “material questions of fact remain” about whether the decedent posed the threat of serious physical harm that Tennessee v. Garner requires before deadly force may be used, id. at 473. The court described the immunity question as “close.” Claims do survive. They survive when a plaintiff can put a specific enough prior case in front of the court.
The circularity problem, in the Supreme Court’s own words
There is a structural objection to all of this, and it does not come from advocates. It comes from the Court.
Pearson v. Callahan, 555 U.S. 223 (2009), freed lower courts from the sequence Saucier v. Katz, 533 U.S. 194 (2001), had imposed — first decide whether a right was violated, then decide whether it was clearly established. Pearson held that “while the sequence set forth there is often appropriate, it should no longer be regarded as mandatory,” and that judges “should be permitted to exercise their sound discretion in deciding which of the two prongs of the qualified immunity analysis should be addressed first.” 555 U.S. at 236. Courts may now grant immunity without ever deciding whether the Constitution was violated.
Pearson itself acknowledged the cost: opinions following the Saucier sequence “often fail to make a meaningful contribution” to the development of constitutional precedent. Id. at 237. Two years later the Court described the resulting loop with unusual candor in Camreta v. Greene, 563 U.S. 692 (2011):
“The court does not resolve the claim because the official has immunity. He thus persists in the challenged practice; he knows that he can avoid liability in any future damages action, because the law has still not been clearly established. Another plaintiff brings suit, and another court both awards immunity and bypasses the claim. And again, and again, and again. So the moment of decision does not arrive. Courts fail to clarify uncertain questions, fail to address novel claims, fail to give guidance to officials about how to comply with legal requirements.”
Id. at 706. Qualified immunity, the Court concluded, “may frustrate ‘the development of constitutional precedent’ and the promotion of law-abiding behavior.” Id. (quoting Pearson, 555 U.S. at 237).
That is the honest account. The rule requires a closely analogous precedent, and the rule itself is part of why closely analogous precedents are scarce.
What it does not reach
Qualified immunity is a defense belonging to individual officials. It is not available to a city or county. Owen v. City of Independence, 445 U.S. 622 (1980), held “that the municipality may not assert the good faith of its officers or agents as a defense to liability under § 1983.” Id. at 638. That does not make municipal claims easier — Monell liability has its own severe requirements, discussed separately here — but it means the immunity analysis and the entity analysis are different questions.
Nor does immunity necessarily end a case that seeks something other than damages. In Lefemine v. Wideman, 568 U.S. 1 (2012) (per curiam), the district court granted the officers qualified immunity and refused nominal damages, yet still permanently enjoined them; the Supreme Court held the plaintiff was a prevailing party entitled to be considered for fees. Id. at 3-4. Injunctive relief and damages are on separate tracks, and so is the fee-shifting question.
What a Minnesotan should take from this
Three things.
The reasonableness of the force is a separate question from immunity, governed by the Graham v. Connor framework. A case can be strong on reasonableness and still end on immunity.
Whether a claim survives depends heavily on Eighth Circuit case law as it stood on the date of the incident — not on how the conduct looks today, and not on decisions from other circuits, which usually will not do the work unless they amount to a robust consensus. Whether such a case exists is a research question, and it should be answered early, because it will determine whether the case is worth bringing.
And this is one more instance of a broader pattern: gates that are decided before anyone reaches the facts, discussed in our piece on access to the civil courts. Understanding that the gate exists is not a reason to abandon a claim. It is a reason to evaluate it seriously and early, with the actual case law in front of you.
Madgett Law, LLC litigates civil rights and police misconduct claims in Minnesota state and federal court, alongside its personal injury and general litigation practice. Qualified immunity is raised early, decided before trial, and immediately appealable when denied — which is why it is worth understanding before a complaint is on file. Send us a message or call 612-470-6529.
Sources: Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982) (objective standard; “insubstantial claims on summary judgment”). Mitchell v. Forsyth, 472 U.S. 511, 526, 530 (1985) (“an immunity from suit rather than a mere defense to liability”; denial of qualified immunity turning on an issue of law is an appealable “final decision” under 28 U.S.C. § 1291). Anderson v. Creighton, 483 U.S. 635, 639, 640 (1987) (particularized definition of the right; the “extremely abstract rights” passage). Ashcroft v. al-Kidd, 563 U.S. 731, 741, 742 (2011) (“beyond debate”; “not to define clearly established law at a high level of generality”; “a robust ‘consensus of cases of persuasive authority’” quoting Wilson v. Layne, 526 U.S. 603, 617 (1999)). Mullenix v. Luna, 136 S. Ct. 305, 308 (2015) (per curiam) (“The dispositive question is ‘whether the violative nature of particular conduct is clearly established.’”). District of Columbia v. Wesby, 138 S. Ct. 577, 590 (2018) (“It is not enough that the rule is suggested by then-existing precedent. The precedent must be clear enough that every reasonable official would interpret it to establish the particular rule the plaintiff seeks to apply.”; the “rare ‘obvious case’” language also at 590). White v. Pauly, 137 S. Ct. 548, 552 (2017) (per curiam) (failure “to identify a case where an officer acting under similar circumstances . . . was held to have violated the Fourth Amendment”). Hope v. Pelzer, 536 U.S. 730, 741 (2002) (“officials can still be on notice that their conduct violates established law even in novel factual circumstances”). Saucier v. Katz, 533 U.S. 194 (2001), and Pearson v. Callahan, 555 U.S. 223, 236, 237 (2009) (sequence no longer mandatory; acknowledgment that such opinions “often fail to make a meaningful contribution” to constitutional development). Camreta v. Greene, 563 U.S. 692, 706 (2011) (block quotation, reproduced verbatim and in full from the run of sentences beginning “The court does not resolve the claim” through “how to comply with legal requirements”; and “may frustrate ‘the development of constitutional precedent’”). Owen v. City of Independence, 445 U.S. 622, 638 (1980) (no good-faith immunity for municipalities). Lefemine v. Wideman, 568 U.S. 1, 3-4 (2012) (per curiam) (officers received qualified immunity; injunction nonetheless conferred prevailing-party status). Eighth Circuit: Kelsay v. Ernst, 933 F.3d 975, 978-80 (8th Cir. 2019) — the panel line reads “Before SMITH, Chief Judge, BEAM, LOKEN, COLLOTON, GRUENDER, BENTON, SHEPHERD, KELLY, ERICKSON, GRASZ, STRAS, and KOBES, Circuit Judges, En Banc,” which is the basis for describing it as en banc; the facts recited above are the court’s own, taken in the light most favorable to the plaintiff, and the court did not decide whether the force was reasonable. Kisela v. Hughes, 138 S. Ct. 1148, 1153 (2018) (per curiam), is cited only as quoted within Kelsay. Ludwig v. Anderson, 54 F.3d 465, 467, 473 (8th Cir. 1995) (reversing qualified immunity for two St. Paul officers; immunity question described as “close”). Tennessee v. Garner, 471 U.S. 1 (1985), is referred to as the standard Ludwig applied and is not independently quoted here. Supreme Court opinions were read from the Caselaw Access Project archive of the United States Reports and the Supreme Court Reporter (static.case.law); Eighth Circuit opinions from the same archive’s Federal Reporter, Third Series. Reporter volume and page come from the archive’s structured citation metadata, and every pin cite above was located in star-paginated text. Cases decided after the archive’s United States Reports coverage ends are cited to the Supreme Court Reporter because official U.S. Reports pagination could not be verified from a primary source in this environment. This article is general legal information about federal law as applied in Minnesota and the Eighth Circuit; it is not legal advice, reading it does not create an attorney–client relationship, and no outcome is promised or implied. Whether qualified immunity applies in any particular case depends on the facts and on the state of Eighth Circuit law when the conduct occurred.